Ligand additivity relationships enable efficient exploration of transition metal chemical space

Abstract

To accelerate exploration of chemical space, it is necessary to identify the compounds that will provide the most additional information or value. A large-scale analysis of mononuclear octahedral transition metal complexes deposited in an experimental database confirms an under-representation of lower-symmetry complexes. From a set of around 1000 previously studied Fe(II) complexes, we show that the theoretical space of synthetically accessible complexes formed from the relatively small number of unique ligands is significantly (ca. 816k) larger. For the properties of these complexes, we validate the concept of ligand additivity by inferring heteroleptic properties from a stoichiometric combination of homoleptic complexes. An improved interpolation scheme that incorporates information about cis and trans isomer effects predicts the adiabatic spin-splitting energy to around 2 kcal/mol and the HOMO level to less than 0.2 eV. We demonstrate a multi-stage strategy to discover leads from the 816k Fe(II) complexes within a targeted property region. We carry out a coarse interpolation from homoleptic complexes that we refine over a subspace of ligands based on the likelihood of generating complexes with targeted properties. We validate our approach on 9 new binary and ternary complexes predicted to be in a targeted zone of discovery, suggesting opportunities for efficient transition metal complex discovery.

Identifying Underexplored and Untapped Regions in the Chemical Space of Transition Metal Complexes

“…To assign the coordination geometry, we categorized the shape of the primary coordination sphere using the coordination number and all possible L 1 -M-L 2 angles. We reported a subset of this data with octahedral geometries (n = 85,575) in prior work 34 , which we call the octahedral mononuclear CSD data set. We also identified a set of unique TMCs by computing their Weisfeiler-Lehman Zatom weighted connectivity graph hashes.…”

“…Because octahedral geometries comprise a significant fraction of TMCs in the mononuclear CSD data set (35%, 85,575 out of 242,829), we carried out an analysis of the symmetry classes 34 present in the octahedral mononuclear CSD data set (Supporting Information Text S1 and Figure S4). We find that TMCs with two identical bidentate ligands in the equatorial plane and two identical axial monodentate ligands (i.e., |22||11|t where "||" groups identical ligands and t indicates ligands trans to each other, as detailed further in Supporting Information Text S1) are the most common symmetry class (Figure 2).…”

Identifying Underexplored and Untapped Regions in the Chemical Space of Transition Metal Complexes

“…To assign the coordination geometry, we categorized the shape of the primary coordination sphere using the coordination number and all possible L 1 −M−L 2 angles. We reported a subset of these data with octahedral geometries (n = 85 575) in prior work, 41 nectivity graph hashes. 42 The CSD primarily contains 3d TMCs with 2p coordinating atoms (Figure 1).…”

“…Because octahedral geometries comprise a significant fraction of TMCs in the mononuclear CSD data set (35%, 85 575 of 242 829), we carried out an analysis of the symmetry classes 41 present in the octahedral mononuclear CSD data set (Text S1 and Figure S4). We find that TMCs with two identical bidentate ligands in the equatorial plane and two identical axial monodentate ligands (i.e., |22||11|t, where || groups identical ligands and t indicates ligands trans to each other, as detailed further in Text S1) are the most common symmetry class (Figure 2).…”

“…To assign the coordination geometry, we categorized the shape of the primary coordination sphere using the coordination number and all possible L 1 –M–L 2 angles. We reported a subset of these data with octahedral geometries ( n = 85 575) in prior work, which we call the octahedral mononuclear CSD data set. We also identified a set of unique TMCs by computing their Weisfeiler–Lehman Atom-wise nuclear-charge-weighted connectivity graph hashes .…”